Home Curriculum Vitae Publications Conference Reports Forthcoming Extramural Colloquia Expert Testimony Teaching Healthcare The Human Ecology of Memory Research Archive Publications and Reports Rants


Does Neuroscience Constrain

 Social-Psychological Theory?

John F. Kihlstrom

 

Note: The following article is based on an exchange that occurred in 2005 on the listserv of the Society for Personality and Social Psychology.  The exchange was prompted by a posting by Prof. Matt Lieberman of UCLA, announcing the publication of Social Cognitive and Affective Neuroscience, a new journal which he edits.  In his message, Lieberman wrote that, in addition to the usual sorts of articles the journal would also publish "companion pieces" that "will strive to make the research more accessible to those who may be interested in the ways in which brain data constrain social psychological theory, but don't want to become neuroscientists themselves" (posting of 11/05/05, emphasis added).  While welcoming the journal, I posted a dissent to the rhetoric of "constraint", which led to an exceptionally interesting exchange of views among listserv members.  I thank Profs. Chris Crandall and Monica Biernat, editors of Dialogue, the SPSP newsletter, for their invitation to prepare a summary of my views, as they evolved from that exchange (Vol. 21, No. 1, 26-27, 32, Spring 2006).  What follows is an expanded version of the published commentary, with references.


Link to French Translation by Anna Chekovsky

Link to Bulgarian translation by Artem Delik


 

One of the most important developments in contemporary social psychology is its embrace of neuropsychological and neuroscientific methodologies (Adolphs, 1999; Cacioppo, Berntson, & McClintock, 2000; Klein & Kihlstrom, 1998; Ochsner & Lieberman, 2001).  This state of affairs has been marked by the publication of special issues of social-psychological journals devoted to neuroscientific research (JPSP, October 2003, e.g., Harmon-Jones & Devine, 2003) and of neuroscientific journals devoted to social-psychological topics (Journal of Cognitive Neuroscience, December 2004, e.g., Heatherton, 2004; NeuroImage, November 2005, e.g., Lieberman, 2005; Brain Research, March 2006, e.g., Beer, Mitchell, & Ochsner, 2006), comprehensive handbooks such as Foundations in Social Neuroscience (MIT Press, 2005), and, beginning later this year, Social Cognitive and Affective Neuroscience -- an entire journal devoted to the new specialty.

I certainly welcome these signs that social psychologists are taking an interest in neuropsychological and neuroscientific methods. At the same time, I want to dissent from a rhetorical stance that appears to be widely shared in neuroscientific circles: the idea that neuroscientific findings can, or will, or should, "constrain" psychological theory. For example, Cacioppo and Berntson (1992) wrote that "knowledge of the body and brain can usefully constrain and inspire concepts and theories of psychological function..." (p. 1025). Similar sentiments were expressed more recently by Ochsner and Lieberman (2001, p. 726).  

Inspire, yes; extend, perhaps; but constrain

If we mean that data from neuropsychological and neuroscientific experiments constrains theory in the same way that data from behavioral and self-report studies constrains theory -- which is to say that theory has to conform to data -- then there is no argument. For example, Tulving (1993) and Klein et al. (1996) obtained data from amnesic patients that supported the hypothesis that episodic (behavioral) self-knowledge was represented in memory independently of semantic (trait) self-knowledge). But the neuropsychological data merely supplemented evidence already available from studies of priming (Klein & Loftus, 1993). Neuropsychological evidence didn't constrain the theory, though it did inform it. It would have been interesting if the patients had lost both episodic and semantic knowledge, but even that fact wouldn't have constrained the hypothesis that the two forms of memory are normally independent.

Taken with its dictionary meaning, which has to do with (en)forcing, compelling, obliging, etc., the use of the term constrain suggests that the neuropsychological level of analysis is somehow privileged -- that neuroscientific evidence is decisive with respect to social-psychological theory. Put bluntly, it betrays the idea that social psychology can't make theoretical progress without neuroscience; that -- paraphrasing and reversing Neisser (1967, p. 1) -- psychology is indeed just something to do until the biochemist comes. This point of view was well expressed by Lucy Brown, a neuroscientist quoted by Benedict Carey in a recent New York Times article on social neuroscience ("Searching for the Person in the Brain", 02/05/06): "Everyone thought phenomena like love and jealousy were simply impossible to study, that they were too variable, too individual. They preferred to think of them as magic." Carey goes on to write: "Imaging and other techniques have now parted the curtain."  

So much for the vast social-psychological literature on interpersonal attraction, friendship, and, yes, love and jealousy (e.g., Berscheid & Walster, 1969, 1978; Buss, 2000, 2003; Rubin, 1973, 1980).

Actually, it has to be said that there does not appear to be any instance where neuroscientific findings have constrained social-psychological theory. Of course, social neuroscience is very young. 

Still, cognitive neuroscience has been around much longer, and it's hard to think of any instance in that field, either (Coltheart, 2006a, 2006b).  For example, sensory physiologists have made great strides in understanding the brain mechanisms involved in color vision, but the basic "opponent process" theory of color vision was worked out in pretty much its modern form more than a century ago by Hering, and revived more recently by Hurvitch and Jameson, on the basis of behavioral evidence (like negative afterimages).  Perhaps the discovery in the 1950s, by Hubel and Wiesel among others, of "bug detectors" and other orientation-specific receptive fields in frogs and cats counts, because it changed our view about how low-level vision works. But there are very few others examples like that.

To the contrary, it appears that precisely the reverse is true: psychological theory constrains the interpretation of neuropsychological and neuroscientific data. My favorite example is the amnesic patient H.M., who put us on the road toward discovering the role of the hippocampus in memory. But what exactly is that role? The fact is, our interpretation of H.M.'s amnesia, and thus of hippocampal function, has changed as our understanding of memory has changed. First, H.M. was thought to have lost his capacity to learn; then to have lost long-term but not short-term memory; then deep but not shallow processing; then declarative but not procedural memory; then episodic but not semantic memory; then explicit but not implicit memory; and now, most recently, relational but not non-relational memory. Here, clearly, neuroscientific data didn't do much constraining: psychological theory changed almost wantonly, while the neural evidence stayed quite constant.

Now, it might be claimed that H.M. did introduce the principle that memory is not a unitary entity. In that way, neuropsychological data would constrain psychological theory, even if further behavioral research were needed to determine exactly what those systems were. Historically, however, the notion of multiple memory systems was already in theory before any neuropsychological or neuroscientific evidence was available.  

In particular, the procedural-declarative distinction popularized by Cohen and Squire (1980) was originally introduced by Winograd (1972, 1975) and Anderson (1976), without the benefit of data from amnesia, and actually goes back even further than that, to the philosophical work of Ryle (1949) and of Bergson (1911).  
Similarly, the episodic-semantic distinction was introduced by Tulving (19872) mostly on logical grounds, but in any event before he ever began at amnesia. 
While the distinction between episodic and semantic memory does, arguably, arise from amnesia work, Schacter's (1987) early review makes clear that the basic idea of implicit memory had been around since roughly Ebbinghaus.  Moreover, it's a simple fact that most subsequent research on explicit and implicit memory -- Roediger's work, for example, or Schacter and Cooper's -- was performed with neurologically intact subjects. So it's hard to say that neuroscientific research played a huge role in, for example, the discovery of a perceptual-representation system, or the articulation of transfer-appropriate processing as a plausible theoretical alternative to multiple memory systems.  

Perhaps the most frequently posited example of neuroscience constraining psychological theory is the debate over the representation of mental imagery -- whether images are represented in propositional or analog form.  I've thought a lot about this example -- in fact, I think it was from Steve Kosslyn that I first heard the "constrains" phraseology, back when we were colleagues at Harvard.  But again, I don't think it works.  

As early as 1980, Ron Finke wrote a Psych Review paper in which he argued, solely on the basis of behavioral evidence, without benefit of any neuroscientific data, and perception and imagery shared common mechanisms.  
And around the same time (1977, 1980, 1981) Kosslyn was making assumptions about the equivalence of imagery and perception in his computer simulation models -- again, based solely on data of human performance, such as his classic experiments on image scanning.  
In 1994, Kosslyn announced the resolution of the "mental imagery debate", citing work like Martha Farah's on visual agnosia and his own brain-imaging work, which showed that perceiving and imaging made engaged the same brain structures.  
Prior to this time, people like John Anderson had argued that the imagery-proposition debate was essentially undecidable -- and in his own work, Anderson didn't hesitate to accept that there were both proposition-based representations and perception-based representations stored in memory.  

So, in this instance, neuroscientific data may have clinched the case.  But I think that pretty much everybody except Zenon Pylyshyn already believed it to be true anyway.  What else could an image be but a percept without a stimulus, constructed from memory?  And if so, where else would images be constructed, other than the same parts of the brain that construct percepts?  So I think that this example is less of a constraint on theory than a proof of a theory that most people already thought highly plausible.  Brain-imaging data complemented the behavioral data, but it didn't constrain the theory of imagery.

Ironically, in any event, Pylyshyn's most recent writings indicate that he's not convinced by the neuroscientific data.    

To be sure, neuroscientific data does constrain neuroscientific theories -- that is, theories about brain structure and function. If you want to know what part of the brain processes memories, evidently you should look around the hippocampus, ruling out structures like the amygdala, and ruling in structures like the parahippocampal gyrus and the entorhinal cortex, which together with the hippocampus comprise a "medial-temporal lobe memory system" (Squire & Zola-Morgan, 1991). But if you want to know the psychological function of that or any other brain structure or system, then you need a well-worked-out theory of memory, and associated behavioral methodologies, already in hand. Thus, the story of H.M. and the hippocampus illustrates quite the opposite of the rhetoric of constraint: that neuroscientific data can be interpreted only within the framework of a valid psychological theory of structure and function.

As a further illustration, consider a neuroscientific claim that lies closer to the interests of social psychologists: that the fusiform area, near the junction of the temporal and occipital lobes, is specialized for processing faces (Kanwisher, McDermott, & Chun, 1997). The claim is based on both neuropsychological analyses of prosopagnosic patients, who appear to suffer a specific deficit in recognizing faces, and brain-imaging data of face-processing by neurologically intact subjects. Such evidence implies that the processing of faces is somehow different from the processing of other, nonsocial, objects. If true, such neuroscientific evidence might indeed constrain psychological theory. But not necessarily: even if different brain systems processed social and nonsocial information, they might do so according to the same principles. More important, there is increasing evidence that the fusiform area is specialized for expert recognition of all sorts of objects at subordinate levels of categorization -- not just faces, but also birds, snowflakes, and greebles (Tarr & Gauthier, 2000). As with the hippocampus, neuroscientific data does not constrain psychological theory, but psychological theory -- in this case, a theory of conceptual structure -- does constrain the interpretation of neuroscientific data.

Both the MTL and the FFA are expressions of the doctrine of modularity that dominates current neuroscientific thinking, and I'm sure that there's some truth to it. However, it should be noted that the earliest proposals for the modularity of language and perception were generally based on behavioral evidence, not to mention introspection, rather then neuroscientific data (Fodor, 1983).  It must also be said that neuroscientific evidence, such as obtained from H.M. and other amnesic patients, isn't necessarily decisive in this respect. Consider, for example, the proposal by Squire and Knowlton (1994) that separate memory systems mediate recognition and categorization. To the contrary, Nosofsky and his colleagues (e.g., 1999) have shown that the amnesia findings can be simulated, in every detail, by a computational model of memory that entails only a single memory system, with different criteria for answering questions about recognition or categorization. So, again, neuroscientific research doesn't constrain psychological theory. If anything, psychological theory -- this time in the form of an operating computer simulation -- offers a plausible alternative to, if not a constraint on, neuroscientific theory. If Nosofsky is correct, memory just might be a "unitary entity" after all. 

And I'm also certain that social cognition differs from nonsocial cognition in important and fundamental ways.  If, as the constructivist approach to perception holds, stimulation is inherently ambiguous, this must be even more true in the social than in the nonsocial case.  Similarly, social perception is likely to be even more influenced by contextual features than nonsocial perception.  Beyond these quantitative differences, though, lies a qualitative one: in social perception, the object of regard is a conscious, sentient being who is actively trying to shape the perceiver's perception of him or her; therefore, the social perceiver has to "read between the lines", and "listen with the third ear", in order to correct for the object's intentional activity.  This is not seem to be the case in nonsocial domains.

And, for that matter, I'm also certain that neuropsychological and neuroscientific methods can shed unique light on certain aspects of social cognition and behavior.  Consider, for example, the amnesic patient H.M. (and others like him).  How in the world does an amnesic person build and maintain social relationships?  Can an amnesic fall in love?  How much does the sharing of memories contribute to fostering close personal relationships?  What happens to existing relationships when memory fails?  Remarkably, such questions have (hardly) ever been asked by the cognitive psychologists who have worked most closely with H.M. and others of his kind (the exception is the science journalist Philip Hilts, in his "biography" of H.M., Memory's Ghost).  One of the goals of Stan Klein and myself, when we wrote our 1996 and 1998 papers advocating social-cognitive neuropsychology, was to get social psychologists interested in these kinds of questions.

Paraphrasing the philosopher John Searle (2001), after you've worked out the problem at the mental level, you can kick it over to the neuroscientists to see how the brain does it. But if the analysis of mental function is wrong, then neuroscience offers little more than souped-up phrenology. So let's do social neuroscience -- but let's not do it in the belief or hope that such evidence will "constrain" our theorizing, or rescue us from whatever theoretical indeterminacies we might suffer. That would be to put the cart before the horse. Good social-psychological theories will make for good social neuroscience. After all, psychology without neuroscience is still psychology; but neuroscience without psychology is just neuroscience.

A note in passing: when Stan Klein and I wrote our 1996 and 1998 papers we pretty deliberately chose to use the term "neuropsychology" as opposed to "neuroscience". Admittedly, "neuropsychology" has a kind of old-fashioned sound to it, conjuring up images of the Bender-Gestalt Test, Halstead-Reitan Neuropsychological Battery, and the Rey-Osterreich Figure, whereas "neuroscience" conjures up images of those machines -- and, well, those images. On the other hand, neuropsychology at least keeps the focus on psychology as the science of mind and behavior, while neuroscience is -- well, the science of neurons.  

 

Epilogue

In a response to my original listserv posting, Matt Lieberman agreed with my general point, but offered a study by Mitchell, Macrae, & Banaji (Journal of Neuroscience, 2004) as an "exception that proves the rule" -- that is, as "a clear example of neuroscience data challenging our theoretical accounts of existing behavioral data."  Lieberman continued (bullet points mine, for clarity):

They replicated the Hamilton paradigm in which people read sentences about target individuals with a goal to memorize the sentences or the goal to for a social impression of the targets.  
The classic finding, of course, is that the social encoding goal leads to better subsequent memory for the sentence than the memorization goal. As far as I am aware, the generally accepted explanation is that this occurred because social encoding leads to deeper encoding than the memorization goal. Importantly, in this account, there is a single encoding process responsible for subsequent memory under both goal instructions, but the process is thought to run better during social encoding than memorization.
Jason's work shows that this account is quite unlikely to be accurate.  They found that while hippocampus activation during memorization instructions predicted subsequent memory for memorized sentences (a result that replicates several previous studies linking hippocampus activity to subsequent memory for non-social information), it did not predict subsequent memory for sentences learned under the social encoding goal.  Instead, activity in dorsomedial prefrontal cortex (a region repeatedly linked to social cognition and theory of mind in previous studies) predicted subsequent memory for sentences learned with the social encoding goal, but not those learned with the memorization goal.
In other words, two different neural systems are at work depending on whether the encoding goal was social or non-social.  This is directly counter to the prevailing account that a single system operates in both cases but that deeper encoding during social encoding leads to more efficient use of this system.  In my opinion, this directly challenges existing theory and suggests that new theory is needed to account for this type of data.  Additionally, these results open up the possibility that the encoding and retrieval rules for social information may differ from the rules for non-social information. This is a very exciting possibility that should be followed up and is likely to provoke new behavioral findings that wouldn't have been produced otherwise.

In response, I expressed the view that while the Mitchell et al. (2004) study is very interesting, I also think that the example is not so clear (again, bullet points for clarity): 

Recall that Hamilton et al. (1980) found that an impression set led to superior memory for behavioral information, compared to a memory set.  Hamilton et al. suggested that impression-formation enhanced memory organization, which in turn enhanced recall.  Another common interpretation of this outcome is that impression-formation leads to deeper, more elaborate processing than does rote memorization.  Both interpretations tacitly assume the operation of a single memory system, though I don't think it's quite right to say that they assume a single encoding process, because traditional memory theory has long made both quantitative and qualitative distinctions among "levels" of processing, and elaboration and organization are arguably distinct forms of memory processing. 
Mitchell et al. performed an experiment "to examine whether social cognition recruits distinct mental operations" (from the first sentence of their abstract).  To this end, they scanned subjects while they were performing an impression-formation task and a control task in which they memorized the sequence in which behavioral statements were presented.  First, they showed that, as predicted, impression-formation led to better memory than sequence-memorization.  But note that the control task was very different from the one employed by Hamilton et al., which simply asked subjects to memorize behavioral statements.  The impression set led to better memory than the memory set, but without the prior Hamilton study, we wouldn't know whether that was because the impression set boosted memory, or whether the memory-set impaired memory by virtue of the seemingly overwhelming requirement to remember the order in which ten (10) items appeared on each of nine (9) different lists.
In any event, Mitchell et al. found that the impression set selectively activated a region of the dorsomedial prefrontal cortex (bilaterally), whereas the sequencing set selectively activated  a region of the right hippocampus.  In light of a review by Gallagher and Frith (2003), which associated the medial prefrontal cortex, along with some other anterior structures, with the "theory of mind", Mitchell et al. suggested that the medial prefrontal cortex is specialized for social cognition.  They further concluded, on the basis of their results, that social cognition doesn't invoke the same brain systems as nonsocial cognition. 
Now, it should surprise nobody that the hippocampus is activated in a straight explicit episodic memory task, and it should surprise nobody that the prefrontal cortex is activated during a task that requires some level of judgment.  But to argue that some brain region is specialized for social cognition requires that the social and nonsocial processing tasks are comparable -- that is, subjects would have to make a social judgment, and a nonsocial judgment, on the same kind of materials.  Mitchell et al. kept the materials constant across their social and nonsocial conditions, but in other respects the two tasks were entirely different.  The social task required subjects to form an impression of each of nine targets' personality traits.  The nonsocial task, as indicated above, required subjects to memorize the order in which behavioral items were presented for each of nine targets.  If we really want to assert the claim that some portion of the prefrontal cortex is specialized for impression formation, we need a control task requiring a nonsocial judgment of equal complexity.  I believe that such a control condition is missing in the Mitchell experiment. 
Just to be clear: In the Mitchell et al. experiment, subjects were asked to make inferences (about personality traits) in the social processing condition; but they weren't asked to make inferences about anything in the nonsocial processing condition -- rather, they were asked to process the order in which items were presented.  In the vocabulary of single-system theories of memory, trait inference is an excellent example of conceptually driven processing; sequence memorization, by contrast, is an excellent example of perceptually driven processing.  As such, the Mitchell et al. experiment confounds social vs. nonsocial processing with conceptual vs. perceptual processing.
I should also note that, although Gallagher and Frith (2003) located the "theory of mind mechanism" in the medial prefrontal cortex (as well as the superior temporal sulcus and the temporal poles), later that same year Saxe and Kanwisher (2003) located that very same mechanism in the temporo-parietal junction -- a posterior brain region which is a long, long way from the anterior regions implicated by Gallagher & Frith.  Saxe and Kanwisher went to some trouble to devise a nonsocial inference task that was comparable to their social inference task, so their results can't simply be dismissed.  I'm not taking a position on where the "theory of mind mechanism" is located.  I don't really care -- precisely because the issue of localization is separate from the question of what the theory of mind is and how it bears on social cognition.  
Of course, it's also possible that reasoning about personality traits (which is what Mitchell et al. asked their subjects to do) is performed by a different brain system than reasoning about mental states (which is what Saxe and Kanwisher asked their subjects to do).  I wouldn't be surprised.
Even if the Mitchell et al. findings were to be sustained by further research, meaning that there's an area of prefrontal cortex that is activated in a social but not in a nonsocial task, it's not at all clear that their findings, as they claim, cast doubt on the applicability of the elaboration and organization principles, derived from research on nonsocial memory, to the case of person memory.  Even if social memory were stored in the prefrontal cortex and nonsocial memory stored in the hippocampus -- and I'm not suggesting for a moment that Mitchell et al. believe this -- both memory systems might obey the elaboration principle derived from studies of nonsocial memory.  But assuming, as most neuroscientists do, that individual memories are represented as reverberating networks of neurons distributed widely across neocortex, the elaborative processing that makes memories highly accessible has to come from somewhere. Arguably, one source of this elaboration might be inferential activity performed by structures in the prefrontal cortex.  Put another way, inferential activity carried out in prefrontal cortex produces a highly elaborate trace of person memory that is stored right alongside nonsocial memories.

I don't mean to pick on the Mitchell et al. (2004) study.  I actually think that social cognition is different from nonsocial cognition in important respects.  But since Matt brought it up, I thought I'd hold my own feet to the fire and work it through.  Far from showing how neuroscientific findings constrain psychological theory, it seems to me that the study actually illustrates my points -- which are that (1) neuroscientific findings are arguably the best way to find out about the biological substrates of a psychological process; but (2) the search for biological substrates assumes that the investigator already knows what is going on with the experimental task at the psychological level of analysis. 

In response, Matt pointed out, correctly, that a subsequent paper by Mitchell, Macrae, & Banaji in NeuroImage (2005b) dealt with the apparent conflict between the Mitchell et al. (2004) findings and those of Saxe and Kanwisher (2003).  The 2005 paper is actually a followup to another study by Mitchell, Heatherton, & Macrae published  in PNAS in 2002.  All three Mitchell studies -- 2002, 2004, and 2005 -- argue that medial prefrontal cortex is specialized for social cognition.  In my earlier message, I pointed out that there were some problems with the 2004 paper that, in my view, vitiated that conclusion.  The 2002 and 2005b are better controlled, in that respect: the 2002 paper compared semantic judgments about people vs. objects, while the 2005 paper compared psychological vs. physical judgments of people vs. dogs.  In both studies, the social judgment differentially activated mPFC.  Further, the 2005b paper has some comments about the Saxe & Kanwisher paper (NeuroImage 2003), which had argued that social judgments were localized in the temporo-parietal junction.  In my earlier message, I did not take a position on who was right in this matter (and I don't actually care).  But note that this is an issue about localization, not about whether neuroscience constrains psychological theory.  Neither the 2002 nor the 2005b paper makes any argument about constraint.  They are solely concerned with localization.  To be clear: I have no problem with the use of neuroscientific methods and data to address issues of functional specialization within the brain.  I can't imagine how we could address the question of the neural bases of social cognition without such data.

Note added July 2006: Additional studies published by Mitchell et al. in 2005 (a) and 2006 add further evidence favoring the role of the mPFC in social cognition.  

The 2005a paper (published after 2005b, but getting the "a" suffix by virtue of order of authorship) scanned subjects while they made either mental or physical judgments of faces: the mentalizing task activated mPFC.
The 2006 paper employed a variant on the procedure of the 2004 paper, in which subjects were scanned while performing either an impression-formation task or an order-memorization task.  As noted above, this comparison risks confounding social nonsocial processing with conceptual vs. perceptual processing.  Interestingly, however, mPFC was activated (somewhat) even in the sequencing condition, when the stimulus materials were trait-diagnostic behavioral statements.  

In any event, these 2006 later studies, like the earlier ones, were more-or-less straightforward exercises in localization.  As such, they addressed issues concerning functional specialization, but did not employ the rhetoric of constraint, and thus does not bear on the issue I raised at the outset, which is whether neuroscientific data can constrain psychological theory.  

Even assuming that Mitchell et al. are right, that social memory is processed by the mPFC, and nonsocial memory is processed by the (right) hippocampus, that neuroscientific fact would not, in and of itself, constrain social-psychological theory in any way.

In the first place, the mPFC and rHipp might obey precisely the same principles, like elaboration, organization, and encoding specificity, even though they process category-specific information.  We drive on the right in America, they drive on the left in Japan, but everybody stops at red lights. 
Moreover, the finding of category-specific functional specialization would not constrain psychological theory even if there were some principle of social memory that did not apply to the nonsocial case (or vice-versa).  This is because this theoretical principle would necessarily be discovered by the usual sorts of behavioral experiments.  If we then discovered that mPFC processed social memories, then we would have to assert that the mPFC instantiated this principle in some way. 
So, if social perceivers must "read between the lines" to make sense of the social world, then the brain system that performs social perception has to be able to do that.  But knowing that mPFC processes social memories doesn't tell us anything about how social memories are processed.  It just tells us where they're processed.  How they're processed is something that has to be discovered by psychological experiments.

Actually, I'm also certain that social cognition differs from nonsocial cognition in important and fundamental ways.  

If, as the constructivist approach to perception holds, stimulation is inherently ambiguous, this must be even more true in the social than in the nonsocial case.  
Similarly, social perception is likely to be even more influenced by contextual features than nonsocial perception.  
Beyond these quantitative differences, though, lies a qualitative one: in social perception, the object of regard is a conscious, sentient being who is actively trying to shape the perceiver's perception of him or her; therefore, the social perceiver has to "read between the lines", and "listen with the third ear", in order to correct for the object's intentional activity.  This is not seem to be the case in nonsocial domains.

And, for that matter, I'm also certain that neuropsychological and neuroscientific methods can shed unique light on certain aspects of social cognition and behavior.  Consider, for example, the amnesic patient H.M. (and others like him).  How in the world does an amnesic person build and maintain social relationships?  Can an amnesic fall in love?  How much does the sharing of memories contribute to fostering close personal relationships?  What happens to existing relationships when memory fails?  

Remarkably, such questions have (hardly) ever been asked by the cognitive psychologists who have worked most closely with H.M. and others of his kind (the exception is the science journalist Philip Hilts, in his "biography" of H.M., Memory's Ghost).  One of the goals of Stan Klein and I, when we wrote our 1996 and 1998 papers advocating social-cognitive neuropsychology, was to get social psychologists interested in these kinds of questions.

Note added April 2006: The rhetoric of constraint continues.  For example, the announcement of a pre-conference workshop on "Probing the Human Mind: Modern Tools for Looking into the Black Box", offered at the annual meeting of the American Psychological Society notes that "The emphasis will be on how models of cognition are constrained and extended by understanding cognitive brain function" Carter & Mangun, 2006).

 

References

Adolphs, R. (1999). Social cognition and the human brain. Trends in Cognitive Sciences, 3(12), 469-479.

Beer, J.S., Mitchell, J.P., & Ochsner, K.N.  (2006).  Special issue: Multiple perspectives on the psychological and neural bases of social cognition.  Brain Research, 1079, 1-3.

Berscheid, E., & Walster, E. H. (1969). Interpersonal attraction. Reading, Ma.: Addison-Wesley.

Berscheid, E., & Walster, E. H. (1978). Interpersonal attraction (2nd ed.). Reading, Ma.: Addison-Wesley.

Buss, D. M. (2000). The dangerous passion: Why jealousy is as necessary as love and sex. New York: Free Press.

Buss, D. M. (2003). The evolution of desire: Strategies of human mating. New York: Basic Books.

Cacioppo, J. T., Berntson, G. G., & McClintock, M. K. (2000). Multilevel Integrative Analyses of Human Behavior: Social Neuroscience and the Complementing Nature of Social and Biological Approaches. Psychological bulletin, 126(6), 829.

Cacioppo, J. T., & Berntson, G. G. (1992). Social psychological contributions to the decade of the brain: Doctrine of multilevel analysis. American Psychologist, 47, 1019-1028.

Carter, C.S., & Mangun, G.R.  (2006, May).  Probing the human mind: Modern tools for looking into the black box.  Pre-conference workshop at the annual meeting of the American Psychological Society, New York.  Described in the "Program Highlights" insert distributed with the April 2006 issue of the APS Observer.

Coltheart, M. (2006a). Perhaps functional neuroimaging has not told us about the mind (so far)? Cortex, 42, 323-331.

Coltheart, M. (2006b). What has functional neuroimaging told us about the mind (so far)? Cortex, 42, 422-427.

Fodor, J. A. (1983). The modularity of the mind. Cambridge, Ma.: MIT Press.

Hamilton, D.L., Katz, L.B., & Leirer, V.O.  (1980).  Cognitive representation of personality impressions: Organizational processes in first impression formation.  Journal of Personality & Social Psychology, 57, 925-939.

Harmon-Jones, E., & Devine, P.G.  (2003).  Introduction to the special section on social neuroscience: Promise and caveats.  Journal of Personality & Social Psychology, 85, 589-593.

Heatherton, T.F.  (2004).  Introduction to special issue on social cognitive neuroscience.  Journal of Cognitive Neuroscience, 16, 1681-1682.

Kanwisher, N. G., McDermott, J., & Chun, M. M. (1997). The fusiform face area: A module in human extrastriate cortex specialized for face perception. Journal of Neuroscience, 17, 4301-4311.

Klein, S. B., & Kihlstrom, J. F. (1998). On bridging the gap between social-personality psychology and neuropsychology. Personality & Social Psychology Review, 2(4), 228-242.

Klein, S. B., & Loftus, J. (1993). The mental representation of trait and autobiographical knowledge about the self. In T. K. Srull & R. S. Wyer (Eds.), Advances in social cognition (Vol. 5, pp. 1-49). Hillsdale, NJ, US: Lawrence Erlbaum Associates, Inc.

Klein, S. B., Loftus, J., & Kihlstrom, J. F. (1996). Self-knowledge of an amnesic patient: Toward a neuropsychology of personality and social psychology. Journal of Experimental Psychology: General, 125(3), 250-260.

Lieberman, M.D.  (2005).  Principles, processes, and puzzles of social cognition: An introduction for the special issue on social cognitive neuroscience.  NeuroImage, 28, 745-756.

Neisser, U. (1967). Cognitive psychology. New York: Appleton-Century-Crofts.

Ochsner, K. N., & Lieberman, M. D. (2001). The emergence of social cognitive neuroscience. AMERICAN PSYCHOLOGIST, 56(9), 717-734.

Mitchell, J. P., Banaji, M. R., & Macrae, C. N. (2005a). The Link between Social Cognition and Self-referential Thought in the Medial Prefrontal Cortex. Journal of Cognitive Neuroscience, 17(8), 1306-1315.

Mitchell, J. P., Cloutier, J., Banaji, M. R., & Macrae, C. N. (2006). Medial prefrontal dissociations during processing of trait diagnostic and nondiagnostic person information. Social Cognitive and Affective Neuroscience, 1, 49-55.

Mitchell, J. P., Macrae, C. N., & Banaji, M. R. (2002). Distinct neural systems subserve person and object knowledge. Proceedings of the National Academy of Sciences, 99, 15238-15243.

Mitchell, J. P., Macrae, C. N., & Banaji, M. R. (2004). Encoding-specific effects of social cognition on the neural correlates of subsequent memory. Journal of Neuroscience, 24, 4912-4917.

Mitchell, J. P., Macrae, C. N., & Banaji, M. R. (2005b). General and specific contributions of the medial prefrontal cortex to knowledge about mental states. NeuroImage, 28, 757-762.

Rubin, Z. (1973). Liking and loving: An invitation to social psychology. New York: Holt, Rinehart, & Winston.

Rubin, Z. (1980). Children's friendships. Cambridge, Ma.: Harvard University Press.

Searle, J. R. (2001). Free will as a problem in neurobiology. Philosophy, 72.

Squire, L. R., & Zola-Morgan, S. (1991). The medial temporal lobe memory system. Science, 253, 1380-1386.

Tarr, M. J., & Gauthier, I. (2000). FFA: A flexible fusiform area for subordinate-level visual processing automatized by expertise. Nature Neuroscience, 3(8), 764-769.

Tulving, E. (1993). Self-knowledge of an amnesic individual is represented abstractly. In T. K. Srull & R. S. Wyer (Eds.), Advances in social cognition (Vol. 5, pp. 147-156). Hillsdale, NJ, USA: Lawrence Erlbaum Associates, Inc.

 

This page last revised 06/13/11 08:25:44 AM.